Welcome to GOVT 10

This class is about teaching you how to:

1. Think about politics scientifically
2. Work with data
3. Thrive in an AI-dominated future

Teaching in the Age of AI

AI has fundamentally changed the world, and we need to work WITH it, not against it.

Some professors are pretending AI doesn’t exist, others are dejected by its rise, but our job is to teach students the skills they actually need.

AI is what you will use in the real world–that is what this class embraces and teaches.

Using Data to Learn About Politics

Different Views of the Same Election

Population-Weighted View

The Data Revolution in Politics

We live in an unprecedented age of political data. Every day, millions of data points are generated about political behavior:

• Voting records - Every vote cast by every legislator
• Campaign finance - Detailed records of political donations
• Public opinion polls - Continuous measurement of citizen attitudes
• Social media activity - Real-time political engagement data
• Election results - Precinct-level voting patterns

Informed Citizenship

Even if you never work in politics professionally, these skills make you a better citizen.

You’ll be able to:

• Critically evaluate claims politicians make with “data”
• Understand what polls really mean (and when to be skeptical)
• Recognize misleading statistics in campaign advertising
• Make sense of election forecasts and their uncertainty
• Participate more effectively in democratic discourse

Political Science is (Sometimes) a Science

Some people question whether political science is really “science.” The answer is that it can be - if we follow the same basic scientific method as other disciplines:

A Historical Example: John Snow

In 1854, in Soho, London was struck by a severe cholera outbreak that killed over 600 people in just a few weeks. Within just 10 days, over 500 people had died, and panic was spreading through the city.

The Conventional Wisdom

Most people believed cholera spread through “bad air” or miasma. Medical authorities were convinced that diseases like cholera were caused by polluted air from rotting organic matter.

Dr. John Snow

Dr. John Snow, a physician and early epidemiologist, challenged the prevailing miasma theory with a radical alternative hypothesis: cholera was transmitted through contaminated water, not airborne particles.

But theoretical disagreement alone was insufficient—Snow needed empirical evidence to overturn established medical doctrine.

The Data-Driven Solution

Snow systematically mapped the data.

• First he placed a mark on a map for every person who died of cholera

The Data-Driven Solution

He then:

• Marked the locations of all public water pumps in the area
• Looked for patterns in the data

What the Data Revealed

Evidence-Based Action

Snow’s evidence was so compelling that local authorities agreed to remove the handle from the Broad Street pump.

The cholera outbreak ended shortly thereafter, saving countless lives.

The Traditional Approach is Becoming Obsolete

The traditional approach is becoming obsolete:

• Memorizing equations
• Debugging code line by line
  
• Struggling with programming details

In 2025, successful data analysts focus on intellectual work:

• Understanding what questions to ask
• Designing appropriate analyses
• Interpreting results correctly
• Communicating findings effectively

This Course Reflects That Reality

We use AI as a powerful assistant that:

• Handles the technical implementation
• Allows us to focus on higher-order thinking that no AI can replace

You’ll learn to be sophisticated consumers and creators of quantitative political analysis, equipped with:

• Technical skills
• Critical thinking abilities

Our Focus: Critical Thinking Over Syntax

Rather than memorizing programming syntax, we focus on:

• Understanding concepts and their applications to real political problems
• Interpreting results and their political implications in context
• Asking the right questions and designing appropriate research strategies
• Verifying logic and critically evaluating AI suggestions for accuracy and appropriateness
• Recognizing limitations of both data and analytical methods
• Communicating findings clearly to diverse audiences

Why This Approach Matters

Traditional Approach vs. Our Approach

%%{init: {'theme':'dark', 'themeVariables': {'primaryColor': 'transparent', 'primaryTextColor': '#ffffff', 'primaryBorderColor': '#4fc3f7', 'lineColor': '#4fc3f7'}}}%%
graph TD
    A1["<b>📚 Memorize<br/>R syntax</b>"]
    A2["<b>🐛 Debug code<br/>errors</b>"] 
    A3["<b>💻 Focus on<br/>programming</b>"]
    A4["<b>⚙️ Struggle with<br/>technical details</b>"]
    
    A1 --> A2 --> A3 --> A4
    
    classDef traditional fill:transparent,stroke:#e6308a,stroke-width:3px,color:#e6308a
    
    class A1,A2,A3,A4 traditional

%%{init: {'theme':'dark', 'themeVariables': {'primaryColor': 'transparent', 'primaryTextColor': '#ffffff', 'primaryBorderColor': '#4fc3f7', 'lineColor': '#4fc3f7'}}}%%
graph TD
    B1["<b>🧠 Understand<br/>statistical concepts</b>"]
    B2["<b>📊 Interpret results<br/>& check logic</b>"]
    B3["<b>❓ Focus on<br/>research questions</b>"] 
    B4["<b>💡 Think critically<br/>about findings</b>"]
    
    B1 --> B2 --> B3 --> B4
    
    classDef modern fill:transparent,stroke:#5ba300,stroke-width:3px,color:#5ba300
    
    class B1,B2,B3,B4 modern

AI-Assisted Analysis

Instead of spending weeks learning to load data, summarize it, and run a statistical model, you will learn how to get the right answer from AI.

And you will learn how to understand what the AI is doing.

Common AI Mistakes to Watch For

AI is powerful but not infallible. You’ll need to watch out for these common mistakes:

Your Job: Critical Thinking

Throughout this course, you’ll develop the judgment to:

• Evaluate whether AI suggestions make sense
• Recognize when results are too good to be true
• Understand the limitations of your data and methods
• Ask follow-up questions when something seems off

Assessment & Grading

Grade Breakdown:

• 25% Midterm Exam (closed book, on paper)
• 25% Final Exam (closed book, on paper)
  
• 30% Quizzes (8 total, 10-15 minutes each, weekly starting Week 2)
• 20% Assignments (weekly submission of class exercises) and Participation (engagement in class and attendance)

Class Format & Technology Rules

During Lecture Portions:

• Closed computer/no devices to maintain focus and engagement
• Take notes by hand - research shows it improves learning

During Hands-On Portions:

• Laptops required for coding exercises and AI-assisted analysis
• You’ll know when to open them!

Required Tools & Materials

Software (All Free/Subscription):

• R (statistical programming language)
• Positron (modern IDE optimized for data science)
  
• Claude Plus subscription (AI assistant for coding and analysis)

Textbook:

• None! I will provide all needed materials
• Everything you need will be on Canvas

Office Hours & Contact

Office Hours: Monday 3:00-5:00 PM EST
Book Appointments: link in the syllabus

Don’t hesitate to reach out if you’re struggling with concepts, coding, or just want to chat about political data!

Our Computing Environment: Positron

We’ll be using Positron, a modern data science IDE that makes working with R and AI seamless.

In Our Next Class

Introduction to R and Data Frames

• Basic R operations with variables and vectors
• Creating and examining data frames
• Loading CSV files using read_csv()
• Essential tidyverse functions for data exploration
• Your first AI-assisted analysis

Final Thoughts

You’ll learn that good quantitative analysis isn’t about having all the answers - it’s about:

• Asking better questions
• Understanding the limits of what we can know

Key takeaways:

• You’ll leave this course not just knowing how to analyze political data
• You’ll understand how to think scientifically about politics in an era of artificial intelligence